Coarse PM Methods Evaluation Study

Coarse PM Methods Evaluation Study Study Design and Preliminary Results Thomas Ellestad, Robert Vanderpool, Paul Solomon, and Mary Harmon USEPA/ORD/NERL Sanjay Natarajan, Chris Noble, and Bob Murdoch Research Triangle Institute Vanderpool.Robert@epa.gov 919-541 541-7877

Background 1997 air regulations established NAAQS for PM2.5 and PM10 as separate metrics U.S. courts have reviewed subsequent litigation and ruled that the PM10 metric is a poorly matched indicator because it includes the PM2.5 fraction EPA has since been considering the possibility of vacating the PM10 regulation and developing a separate standard for PMc

Study Objectives Evaluate the field performance of leading methods for monitoring the coarse fraction of PM10 (PMc = PM10 PM2.5) Evaluate samplers which are either already commercially available or in their final stages of development Include both filter-based (time-integrated) integrated) and semi-continuous measurement methods

PM2.5 and PM10 FRM Samplers PM10 (WINS Removed) PM2.5 (with WINS) P M 10 PMc = PM10 PM2.5 Standard low-vol PM10 inlets aspirating at 16.7 lpm (actual conditions) PM2.5 aerosol fractionation using awins equipped with DOS impaction oil Filters were conditioned at 22C and 35% RH, analyzed gravimetrically. Post- sampling filters archived at -30C for subsequent chemical analysis 3 FRM pairs from BGI, R&P, and Thermo-Andersen equipped with teflon filters (4 th FRM pair equipped with quartz filters)

R&P Partisol-Plus Plus 2025 Dichot Standard PM10 inlet aspirating at 16.7 lpm (actual) Aerosol fractionation by custom virtual impactor (15 lpm and 1.67 lpm) PM2.5 and PMc mass collected on 47 teflon filters for gravimetric analysis Sequential sampler with multi-day capability 4 units used in our study (3 teflon and 1 quartz)

R&P Coarse Particle TEOM Modified PM10 inlet aspirating at 50 lpm (actual) PM10 aerosol is fractionated by a custom virtual impactor (2 lpm coarse flow and 48 lpm fine flow) PMc fraction is heated to 50 C to remove particle bound water Coarse aerosol is collected and quantified by a standard TEOM sensor 3 units used in our study

Tisch SPM-613D Dichot Beta Gauge Standard PM10 inlet aspirating at 16.7 lpm (~std) Aerosol heated >25C Aerosol fractionation by custom virtual impactor PM2.5 and PMc mass collected on polyflon tape roll PM2.5 and PMc mass quantified hourly using separate beta sources and detectors 3 units used in our study

TSI Model 3321 Aerodynamic Particle Sizer Standard PM10 inlet aspirating at 16.7 lpm (actual) Isokinetic fraction of PM10 aerosol removed at 5 lpm and enters the APS inlet APS sizes individual particles aerodynamically using time of flight approach Single particle volume converted to mass using mean density provided by user Total aerosol mass is sum of individual particle mass APS provide only PMc; not applicable for PM2.5 or PM10 Only sampler in study which provides detailed size distribution information 2 units used in our study

Mobile Sampling Platform (Side View)

Mobile Sampling Platform (Front View)

Sampler Performance Issues Relative bias versus collocated FRMs Precision (2 or 3 samplers of each type) Field reliability Evaluation under a wide range of weather conditions and aerosol types

QA/QC Initiatives QAPP was reviewed and approved by EPA Study design and operation passed EPA s systems audit SOPs were reviewed by the sampler manufacturers Sampler manufacturers were allowed to verify the working condition of their respective samplers prior to sampling at each site Sampling and fractionation components cleaned prior to each study NIST-traceable sampler calibration equipment was used for all sampler calibrations and audits Three performance audits and three field blank tests were conducted at each site Replicate weighings were conducted at the site as well as at EPA s RTP weighing facility

Study Details Using 22 hour, daily sampling periods for comparisons (11 am to 9 am local time) Chemical analysis (XRF, IC, thermal optical) of archived filters will provide particle composition, which may explain observed sampler performance Full data set from all three sites will be ready by Feb 2004. Detailed results from Gary, IN will be presented at the October AAAR meeting

Study Sites RTP, NC (10 days of shakedown tests, Jan. 2003) Gary, IN (30 days of tests under cold, snow/rain, variable PM2.5/PM10 ratios, March-April, 2003) Phoenix, AZ (30 days of tests under hot, dusty conditions, consistently low PM2.5/PM10 ratios, May-June, 2003) Riverside, CA (30 days of tests under warm conditions, higher PM2.5/PM10 ratios than Phoenix, July-August, 2003)

Gary, IN

GARY, IN SIZE DISTRIBUTION DATA March - April, 2003 1.00 PM2.5/PM10 Concentration Ratio 0.90 0.80 0.70 0.60 0.50 0.40 0.30 0.20 0.10 0.00 GARY, IN FRM DATA PM2.5/PM10 Range = 0.32 to 0.83; Mean = 0.55 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Sample Day

PMc FRM MEASUREMENTS - GARY vs RTP WEIGHING Gary, IN (March - April, 2003) 60.0 GARY WEIGHING PMc Conc. (micrograms/m 3 ) 50.0 40.0 30.0 20.0 10.0 RTP/Gary = 1.00 RTP WEIGHING 0.0 0 5 10 15 20 25 30 35 SAMPLE DAY

INTERMANUFACTURER PM2.5 FRM MEASUREMENTS (RTP WEIGHING) Gary, IN (March - April, 2003) 60.0 50.0 Max PM2.5 = 46.9 ì g/m 3 Min PM2.5 = 10.3 ì g/m 3 PM2.5 Conc. (micrograms/m 3 ) 40.0 30.0 20.0 Mean PM2.5 = 22.8 ì g/m 3 INTERMANUFACTURER CV = 1.5% AND R&P BGI 10.0 0.0 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 SAMPLE DAY

INTERMANUFACTURER PM10 FRM MEASUREMENTS (RTP WEIGHING) Gary, IN (March - April, 2003) 100.0 90.0 PM10 Conc. (micrograms/m 3 ) 80.0 70.0 60.0 50.0 40.0 30.0 20.0 Max PM10 = 84.9 ìg/m 3 Min PM10 = 22.6 ìg/m 3 Mean PM10 = 42.6 ìg/m 3 AND R&P BGI 10.0 INTERMANUFACTURER CV = 2.4% 0.0 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 SAMPLE DAY

INTERMANUFACTURER PMc FRM MEASUREMENTS (RTP WEIGHING) Gary, IN (March - April, 2003) 70.0 60.0 Max PMc = 58.1 ì g/m 3 Min PMc = 4.5 ì g/m 3 Mean PMc = 19.9 ì g/m 3 PMc Conc. (micrograms/m 3 ) 50.0 40.0 30.0 20.0 INTERMANUFACTURER CV = 5.7% AND R&P BGI 10.0 0.0 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 SAMPLE DAY

DICHOT AND FRM TIMELINE (PM 2.5 ) GARY, IN (MARCH - APRIL, 2003) PM 2.5 Conc. (micrograms/m 3 ) 50.0 45.0 40.0 35.0 30.0 25.0 20.0 15.0 10.0 5.0 0.0 DICHOT/FRM = 0.99 PM2.5 FRM R&P DICHOTS 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 Sample Day

Dichot versus FRM PM2.5 Concentrations Gary, IN (March - April, 2003) 50.0 Dichot PM2.5 Concentrations (micrograms/m3) 45.0 40.0 35.0 30.0 25.0 20.0 15.0 10.0 Dichot = 0.99*FRM + 0.0 R Square = 0.998 5.0 0.0 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 FRM PM2.5 Concentrations (micrograms/m3)

DICHOT AND FRM TIMELINE (PMc) Gary, IN (March - April, 2003) PMc Conc. (micrograms/m 3 ) 70.0 60.0 50.0 40.0 30.0 20.0 10.0 Dichot/FRM = 0.89 FRM PMc (PM10-PM2.5) R&P DICHOTS Dichot = 0.87*FRM + 0.4 R square = 0.969 0.0 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 Sample Day

DICHOT AND FRM TIMELINE (PM10) Gary, IN (March - April, 2003) PM10 Conc. (micrograms/m 3 ) 90 80 70 60 50 40 30 20 10 0 Dichot = 0.95*FRM - 0.5 R square = 0.981 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 Sample Day PM10 FRM R&P DICHOTS Dichot/FRM = 0.94

TISCH SPM-613D AND FRM TIMELINE (PM 2.5 ) Gary, IN (March - April, 2003) PM2.5 Conc. (micrograms/m 3 ) 70 60 50 40 30 20 10 Tisch = 1.17*FRM + 1.6 R square = 0.948 TISCH CV = 7.1% PM2.5 FRM TISCH SPM-613D TISCH/FRM = 1.26 0 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 Sample Day

TISCH SPM-613D AND FRM TIMELINE (PM c ) Gary, IN (March - April, 2003) PMc Conc. (micrograms/m 3 ) 70 60 50 40 30 20 10 0 TISCH/FRM = 0.91 TISCH = 0.89*FRM + 0.3 R square = 0.979 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 Sample Day FRM PMc (PM10-PM2.5) TISCH SPM-613D TISCH CV = 10.5%

R&P COARSE TEOM AND FRM TIMELINE (PMc) Gary, IN (March - April, 2003) PMc Conc. (micrograms/m 3 ) 70 60 50 40 30 20 10 TEOM PMc/FRM = 0.69 PMc (PM10-PM2.5) TEOM = 0.68* FRM + 0.18 R square = 0.982 R&P CONTINUOUS COARSE 0 TEOM PMc CV = 4.4% 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 Sample Day

TSI APS vs FRM PMc Concentrations Gary, IN (March - April) 70.0 PMc Conc. (micrograms/m 3 ) 60.0 50.0 40.0 30.0 20.0 APS/FRM = 0.42 APS = 0.48*FRM -1.4 R square = 0.795 APS CV = 16.8% APS 1 PMc APS 2 PMc Mean FRM PMc 10.0 0.0 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 SAMPLE DAY

Phoenix versus RTP FRM Weighing May - June 2003 250.0 Mass Concentration (micrograms per cubic meter) 200.0 150.0 100.0 50.0 0.0 PMc PM10 Inter-manufacturerer PM2.5 CV = 3.4% Inter-manufacturerer PMc CV = 3.3% Inter-manufacturerer PM10 CV = 3.6% PM2.5/PM10 Range = 0.10 to 0.28: Mean = 0.22 PM2.5 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 PM2.5 (RTP) PM10 (RTP) PMc (RTP) PM2.5 (Site) PM10 (Site) PMc (Site) Sampling Day

Dichot PM2.5 versus FRM PM2.5 Phoenix, AZ May - June, 2003 Dichot PM2.5 (micrograms per cubic meter) 30.0 25.0 20.0 15.0 10.0 5.0 0.0 Mean Dichot/FRM = 1.09 Dichot = 1.24 * FRM - 1.6 R Square = 0.97 0.0 5.0 10.0 15.0 20.0 25.0 FRM PM2.5 (micrograms per cubic meter)

DICHOT AND FRM TIMELINE (PMc) Phoenix, AZ (May - June, 2003) 250.0 PMc Conc. (micrograms/m 3 ) 200.0 150.0 100.0 50.0 Dichot = 0.70*FRM + 5.0 R square = 0.98 Mean Dichot/FRM = 0.80 PMc (PM10-PM2.5) R&P DICHOTS 0.0 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 Sample Day

DICHOT AND FRM TIMELINE (PM10) Phoenix, AZ (May - June, 2003) 250.0 PM10 Conc. (micrograms/m 3 ) 200.0 150.0 100.0 50.0 Dichot = 0.75*FRM + 5.9 R square = 0.98 Mean Dichot/FRM = 0.84 PM10 FRM R&P DICHOTS 0.0 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 Sample Day

Tisch & FRM PM2.5 Concentrations Phoenix AZ: May - Jun, 2003 50 PM2.5 concentrations (ug/m3) 40 30 20 10 Average Tisch CV = 5.9 % Mean Tisch/FRM = 1.70 Tisch = 2.03*FRM - 3.4 R square = 0.947 Tisch PM2.5-1 Tisch PM2.5-2 Tisch PM2.5-3 Mean FRM-PM2.5 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Sample Day

Tisch, & FRM PMc Concentrations Phoenix AZ: May - Jun, 2003 PMc concentrations (ug/m3) 220 200 180 160 140 120 100 80 60 40 20 0 Average Tisch PMc CV = 9.5 % Mean Tisch/FRM = 1.04 Tisch = 0.92*FRM + 6.0 R square = 0.996 Tisch PMc-1 Tisch PMc-2 Tisch PMc-3 Mean FRM-PMc 1 11 13 15 21 23 25 3 5 7 9 17 19 27 29 Sample Day

TEOM, & FRM PMc Concentrations Phoenix AZ: May - Jun, 2003 PMc concentrations (ug/m3) 220 200 180 160 140 120 100 80 60 40 20 0 TEOM PMc1 TEOM PMc2 TEOM PMc3 FRM-PMc Average TEOM PMc CV = 6.6 % Average TEOM/FRM Ratio = 1.05 TEOM = 0.79*FRM +12.8 R square = 0.951 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Sample Day

APS, & FRM PMc Concentrations Phoenix AZ: May - Jun, 2003 220 200 PMc concentrations (ug/m3) 180 160 140 120 100 80 60 Average APS PMc CV (Days 1-15) = 2.2 % Average APS/FRM = 0.55 APS = 0.56*FRM -0.2 R square = 0.992 APS 1-PMc APS 2-PMc FRM PMc 40 20 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Sample Day

PM Size Distributions (TSI APS) Gary,IN and Phoenix,AZ Delta Mass/Delta Log Dp 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 Gary Phoenix 0 1 10 100 Log Dp

Summary of Results (independent of site) FRMs show strong inter-manufacturer precision (CV<4% for all three metrics) with no tendency for producing negative PMc values Filter-based dichots show strong precision (CV<4% for all metrics) Site weighing results agree closely with RTP results Precision of the semi-continuous samplers is considered to be acceptable Correlation (R 2 ) of all continuous samplers is typically strong versus the collocated FRMs

SUMMARY OF SITE RESULTS GARY, IN PHOENIX, AZ SITE AEROSOL PMc Mean ( ìg/m 3 ) PM2.5/PM10 Range 19.9 0.32-0.83 55.6 0.10-0.28 PM2.5/PM10 Ratio 0.55 0.18 DICHOTS Dichot/FRM PM2.5 1.00 1.09 Dichot/FRM PM10 0.94 0.84 Dichot/FRM PMc 0.90 0.79 TEOM PMc TEOM PMc/FRM 0.69 1.05 TISCH Tisch/FRM PM2.5 1.26 1.70 Tisch/FRM PM10 1.09 1.16 Tisch/FRM PMc 0.91 1.04 TSI APS APS/FRM PMc 0.42 0.55

Future Work Complete RTP gravimetric analysis of Riverside, CA filters (>1500 filter weighings per site) Conduct chemical analysis of archived site filters; potentially use results as explainers of sampler performance Possibly conduct comprehensive field tests at an additional field site Possibly perform laboratory tests with samplers to better understand aerosol fractionation and/or particle loss issues

Acknowledgements Rupprecht & Patashnick, Inc. Tisch Environmental, Inc. TSI, Inc. Indiana Department of Environmental Management (Gary) Maricopa County Environmental Services Department (Phoenix) University of California Ag Ops (Riverside)

Disclaimer The United States Environmental Protection Agency through its Office of Research and Development funded and managed the research described here under Contract 68-D- 00-206. It has been subjected to Agency review and approved for publication.